BACKGROUND OF THE INVENTION

The invention relates to a method of whole-cell catalysed biotransformation of linear alkanes to oxygenated products in actively dividing cells capable of expressing a biocatalyst that catalyses conversion of a linear alkane into an oxygenated product, comprising incubating the actively dividing cells in a biotransformation medium comprising a linear alkane, thereby catalysing the conversion of the linear alkane into the oxygenated product.

In particular, the invention relates to a method of whole-cell catalysed biotransformation of linear alkanes to oxygenated products in actively dividing cells capable of expressing a biocatalyst that catalyses conversion of a linear alkane into an oxygenated product, comprising (i) incubating actively dividing cells in a growth medium including (a) a polysaccharide and (b) a polysaccharide-hydrolysing enzyme that hydrolyses the polysaccharide into a growth substrate for the actively dividing cells at a controlled rate and (ii) incubating the actively dividing cells in a biotransformation medium comprising a linear alkane, thereby catalysing the conversion of the linear alkane into the oxygenated product.

The linear alkanes from expanding activities in the petrochemical industry potentially provide an inexpensive hydrocarbon feedstock for the production of high value oxygenated products such as alcohols, ketones, aldehydes, hydroxyacids, dicarboxylic acids, and the like by means of biotransformation. The application of whole-cells as biocatalysts in biotransformation is the preferred approach to guarantee the continuous regeneration of reducing equivalents from cofactors such as NAD(P)H. It also ensures the structural organisation and stable environment required for the biotransformation reaction in organic solvents (Duetz et al., 2001). In living cells, however, various cellular reactions compete for such cofactors, including oxidative phosphorylation. This has led to the separation of the cellular growth phase from the bioconversion phase during biotransformation processes, through the use of resting cells rather than growing (or actively dividing) cells in order to uncouple the energy demand for cell division from that required for bioconversion. The resting cells are harvested from a growth medium, washed and resuspended in a buffer solution that does not support growth of the cells. Resting cells are therefore distinguished from growing or actively dividing cells in that, although they are metabolically active, their metabolism is not directed towards growth and reproduction. Resting cells have been reported to show higher specific activity and productivity compared with growing cells as the biocatalyst efficiency in growing cells is generally compromised due to formation of by-products which serve as a carbon sink and possibly as a result of limited cofactor availability.

However, the use of resting cells suffers a number of drawbacks, including low catalytic stability and decreasing concentration of active enzyme arising from changes in the regulation of gene expression. This, in turn, results in a lower product titre and limited productivity. These factors (i.e. biocatalyst activity and stability) are important for the economic feasibility of the bioconversion of alkanes to oxygenated products on a process scale.

Biotransformation reactions are usually performed in a batch process. The batch operation is a discontinuous or two-phase process, where there is first a growth phase in which the cells are grown to the desired biomass concentration, followed by a separate bioconversion or biotransformation phase. In the biotransformation phase resting cells that have been harvested from the growth phase are provided with a bioreaction mixture consisting of the resting cell suspension (biocatalyst), carbon and an energy source to enable co-factor regeneration, but not active cell division (aqueous phase), as well as an alkane substrate (organic phase) and incubated for a period of time for the biotransformation process. These reactants are initially introduced into the reaction vessel and operate as a partially closed system, except for exchange of air and pH control agents. The batch reaction is agitated to promote oxygen and substrate mass transfer in the two-phase biocatalytic system. However, the batch system suffers from disadvantages such as high initial concentration of the carbon source and toxicity that may arise from the high concentration of the organic substrate at the start of biotransformation. This encourages uncontrolled process resulting in overflow metabolism, carbon wastage and biocatalyst inhibition.

The fed-batch mode of operation provides a means of controlling cell metabolism of the resting cells. Here the growth-limiting substrate (usually the carbon source) is supplied into the reactor at a predetermined rate which allows better control of cell metabolism. This has been shown for whole-cell-based CYP153A6-catalysed limonene hydroxylation in a 3 L bioreactor, equipped with agitation, temperature and pH control (Cornelissen et al., 2013). Glucose was constantly supplied at a rate of 0.15 g min ^-1 to provide carbon and energy for the cell metabolism (but not active cell division), under biotransformation conditions. A similar feeding regime has been demonstrated by Vallon et al. (2013) in the production of octanol from octane with the use of a 1.5 L bioreactor. However, even with the use of the fed-batch mode of operation, the biocatalyst efficiency obtained for the production of octanol from octane in this study was only 0.016 goctanoi gDcw ^"1 and the biocatalyst activity described was only 1.63 μηιοΙ gpcw ^"1 min ^-1 . Furthermore, the application of fed-batch mode of operation is impracticable in small reaction vessels on shaken platform which are typically used in process development studies.

It would be useful if a more efficient biotransformation method could be developed; in particular a system that could be used both during the process development phase and during the scale-up phase of biotransformation. It would furthermore be useful if such a process was able to be performed in a single bioreactor, thereby reducing the complexity and cost of the unit operations required.

SUM MARY OF THE INVENTION

According to a first aspect of the invention there is provided a method of whole-cell catalysed biotransformation of linear alkanes to oxygenated products in actively dividing cells capable of expressing a biocatalyst that catalyses conversion of a linear alkane into an oxygenated product, comprising incubating the actively dividing cells in a biotransformation medium comprising a linear alkane, thereby catalysing the conversion of the linear alkane into the oxygenated product.

In a preferred embodiment of the invention, the method of whole-cell catalysed biotransformation of linear alkanes to oxygenated products in actively dividing cells capable of expressing a biocatalyst that catalyses conversion of a linear alkane into an oxygenated product, comprises: (i) incubating actively dividing cells in a growth medium including:

(a) a polysaccharide, and

(b) a polysaccharide-hydrolysing enzyme wherein the enzyme hydrolyses the polysaccharide into a substrate for growth of the actively dividing cells at a controlled rate; and

(ii) incubating the actively dividing cells in a biotransformation medium comprising a linear alkane, thereby catalysing the conversion of the linear alkane into the oxygenated product.

More particularly, the method comprises the steps of:

(i) providing cells capable of expressing a biocatalyst that catalyses conversion of a linear alkane into an oxygenated product;

(ii) introducing the cells of (i) into a growth medium that supports active cell division including (a) a polysaccharide and (b) a polysaccharide-hydrolysing enzyme wherein the enzyme hydrolyses the polysaccharide into a substrate for growth of the actively dividing cells at a controlled rate and maintaining active cell division at a controlled rate in the growth medium;

(iii) expressing the biocatalyst in the actively dividing cells of (ii);

(iv) adding a biotransformation medium comprising the linear alkane to the actively dividing cells of (iii);

(v) incubating the actively dividing cells of (iv), thereby catalysing the linear alkane into an oxygenated product; and

(vi) recovering the oxygenated product of (v).

Preferably, at least steps (ii) to (vi) are performed in a single bioreactor.

The method may comprise a further step (iA), wherein prior to step (ii), the cells of step (i) are cultured in an initial growth medium that supports active cell division.

The initial growth medium may be a conventional growth medium known to those skilled in the art), or a growth medium including (a) a polysaccharide and (b) a polysaccharide- hydrolysing enzyme wherein the enzyme hydrolyses the polysaccharide into a substrate for growth of the actively dividing cells at a controlled rate.

The initial growth medium and the growth medium of step (ii) may have a similar composition, apart from the growth medium of step (ii) comprising any one or more of: a different concentration of the polysaccharide-hydrolysing enzyme, a growth-limiting nutrient, or an inducing agent for enzyme expression, compared with the initial growth medium. The step (ii) may comprise a step of inoculating the cells cultured in step (iA) into the growth medium of step (ii). Alternatively, step (ii) may comprise a step of supplementing the cells cultured in step (iA) with growth medium components to produce the growth medium of step (ii).

It is to be appreciated that, although not a preferred embodiment of the invention, a person skilled in the art may, rather than by use of a polysaccharide and polysaccharide-hydrolysing enzyme to control the rate of growth of the actively dividing cells, use a fed-batch system where a substrate for growth of the actively dividing cells is fed into the system at a controlled rate, thereby to control the rate of growth of the actively dividing cells during the method.

The oxygenated product may be an alcohol, a ketone, an aldehyde, a hydroxyacid, a dicarboxylic acid or the like.

The polysaccharide may include starch, amylose, amylopectin, dextrin, cellulose, or hemicellulose and derivatives thereof known to those skilled in the art.

The growth substrate may be sucrose, fructose, glucose or another readily assimilable compound released from the polysaccharide including maltose, or maltotriose. Preferably, the growth substrate is glucose.

The polysaccharide-hydrolysing enzyme may be a single enzyme or a cocktail of enzymes. Typically, the enzyme would be selected depending on the polysaccharide used and the metabolically active growth substrate desired. For example, the enzyme may be an amylase, a glucoamylase, an isoamylase, a beta-glucosidase, a cellulolytic enzyme, or others known to those skilled in the art. Preferably the enzyme is a glucoamylase.

The linear alkane may be an n-alkane selected from any one or more of Ci to C36 alkanes.

In particular the n-alkane may be octane and the octane may be converted into 1-octanol.

The biocatalyst may be an oxygenase. Preferably, the biocatalyst is a hydroxylase. In particular, the biocatalyst is AlkB or cytochrome P450. For example, the cytochrome P450 is CYP153, CYP102, or CYP 52. Most preferably, the cytochrome P450 is CYP153A6.

The cell may be a bacterial or a fungal cell, including a yeast cell.

For example, the bacterial cell may be Escherichia. coli, Pseudomonas sp. , including Pseudomonas putida, Rhodococcus sp. , Act netobacter sp., Bacillus sp. , including Bacillus megate um, Mycobacterium sp, Arthrobacter sp., Streptomyces sp., Marinobacter aquaelolei, or Geobacillus thermodenitrificans.

The fungal cell may be a yeast cell, including Candida sp., Yarrowia or Arxula.

The cell may further express one or more electron transfer protein(s) for providing reducing equivalents for biotransformation. For example, the electron transfer protein(s) may be Ferrodoxin or Ferrodoxin Reductase, or both, or Rubredoxin, or Rubredoxin Reductase or both, or Cytochrome P450 Reductase, or FAD/FMN Reductase.

The cell may further express a transport protein for transporting hydrophobic substrates or products or both across the cell membrane. For example, the transport protein may be AlkL.

The cell may further express or over-express a dehydrogenase enzyme, including a glucose dehydrogenase or glycerol dehydrogenase enzyme.

The cell may be genetically engineered to express any one or more of the biocatalyst, the electron transfer protein(s), the transport protein, or the dehydrogenase enzyme. In particular, the genetically engineered cell may be Escherichia coli.

The method at step (iii) may further comprise inducing the cells to express the biocatalyst. For example, the cells may be induced by addition of Isopropyl β-D-l-thiogalactopyranoside (IPTG), n-alkane or lactose. Preferably, the cells are induced by addition of IPTG or lactose.

The growth medium of step (ii) may comprise a booster solution comprising complex additives, yeast extract and tryptone.

The growth medium of step (iv) may be formulated to exclude complex additives.

The incubation of step (v) may be performed at from about 20 °C to about 37 °C, preferably from about 20 °C to about 25°C and most preferably at about 20 °C. For example, the steps (iii) and (iv) may be performed at a temperature of about 37 °C and the incubation step (v) at a temperature of from about 20 °C to about 25°C, preferably about 20 °C.

The biotransformation medium may further comprise an organic solvent to serve as a product sink, wherein during step (vi) the oxygenated product is drawn out of the aqueous phase into the product sink for recovery.

For example, the solvent may be 1-hexadecanol or bis(2-ethylhexyl) phthalate (BEHP). Preferably, where the oxygenated product is octanol, the solvent is BEHP. According to a further aspect of the invention there is provided a method of whole-cell catalysed biotransformation of linear alkanes to oxygenated products according to the invention, wherein at least from step (ii) onward of the method is performed on a small scale in a vial, a microtitre plate, or the like.

According to a further aspect of the invention there is provided a method of whole-cell catalysed biotransformation of linear alkanes to oxygenated products according to the invention, wherein at least from step (ii) onward of the method is performed in a bioreactor such as a stirred tank reactor or an orbital shaking reactor, or others known to those skilled in the art. The orbital shaking reactor may comprise a helical track. Alternatively, the entire method may be performed in a bioreactor.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 shows Octane bioconversion using metabolically active resting cells and growing whole cell biocatalysts cultured on chemically defined glucose medium. For the use of resting cells, the biomass was resuspended (to 3.5 gDcw LBRM ^"1 ) in 200 mM sodium phosphate buffer (pH 7.2) consisting of glucose. In the growing system, the same biomass was resuspended in glucose-based defined medium (pH 7.2) to a starting concentration of 2.5 gDcw LBRM ^"1 . During biotransformation, the vials were opened intermittently for glucose addition and also to allow inlet of air to avoid oxygen limitation. At specified intervals, the octanol (A), P450 concentration (B), acetate formation (C) and reaction pH (D) were determined.

Figure 2 shows the influence of glucoamylase activity on glucose release from starch polymer in terms of (a) resultant glucose concentration and (b) glucose release rate using the EnBase® medium. The data was collected in the absence of biomass addition. The glucoamylase concentration was in the range 0.6 to 12 U L ¹ . The data represents an average of values from studies with and without octane.

Figure 3 shows the initial rate of glucose release from the polysaccharide substrate through varying the concentration of gluco-amylase concentration. The reaction was carried out in the presence of octane over the duration of 10 h. Figure 4 shows the continuous supply of carbon and energy source in whole-cell hydroxylation of n-octane through the activity of glucoamylase in a starch hydrolysed medium. The (a) octanol formation (b) and pH of the reaction mixture were determined. The biomass was cultured on EnBase® medium, harvested by centrifugation and resuspended in starch solution to a concentration of 6.95 gDcw L ^"1 . The bio-reaction mixture was supplemented with glucoamylase in the range of 0.6 to 12 U L ¹ and 1 ml of the broth was introduced into 60 ml reaction vials, followed by the addition of 200 μΙ octane and 100 μΙ BEHP. The reaction was mixed on an orbital shaker at 200 rpm and 20°C and the vials were opened intermittently to allow air inlet to prevent oxygen limitation.

Figure 5 shows whole cell biotransformation with growing biocatalyst cultured on

EnBase® medium consisting of the complex booster component. The glucoamylase concentration was 0.6 U L ¹ while process temperature was varied from 20 to 30°C, and compared with growing cell on chemically defined medium.

Figure 6 shows the effect of glucoamylase concentration across the range 0.6 - 12 U

L ^"1 on biomass accumulation shown on (a) a linear and (b) a logarithmic scale. The reaction was carried out in 60 ml vial using 1 ml of cell broth (2% v/v inoculum), and incubated at 20°C for 48 h.

Figure 7 shows octane oxidation as a function of glucose release rate, varied by varying the glucoamylase concentration. Following 22 h of cell growth and 1 h of protein induction, 1 ml of cell broth was introduced into 60 ml vials and glucoamylase was added. Thereafter, octane and BEHP were added for biotransformation. The reaction duration was 9 h.

Figure 8 shows (a) whole cell oxidation of octane using growing biocatalyst on

EnBase® medium and (b) the production of acetic acid during biotransformation. E. coli was grown on EnBase® medium for 12 h at 30°C using starting enzyme concentration of 0.6 U L ¹ prior to induction. Following 3 h of induction, 1 ml aliquots were distributed into 60 ml reaction vessel with glucoamylase added in the range 0.6 to 12 U L ^"1 . Thereafter, 200 μΙ of octane and 100 μΙ of BEHP (product sink) were added to commence biotransformation at 20°C and agitation at 200 rpm. Figure 9 shows (a) the oxidation of n-octane using growing whole cell E. coli (expressing CYP153A6) in EnBase® Flo Zero. The biomass was grown on EnBase® for 22 h, induced and incubated at 20°C for 3 h, harvested and re- suspended (1-1.5 gDCW L ^"1 ) in EnBase® Flo Zero supplemented with IPTG, δ-ALA and thiamine. A 1 ml aliquot of this mixture was introduced into the 60 ml amber vials, glucoamylase was added at varied concentration and 30% v/v organic phase consisting of 200 μΙ octane and 100 μΙ BEHP was included to start biotransformation; and (b) the biomass formation. Overflow metabolism was monitored via (c) acetic acid determination followed by (d) detection of medium pH (e) and the excess (un-used) glucose was determined in the reaction mixture. shows OTR characterization of the STR. Error bars reflect the standard error of the mean at the 95% confidence interval (n = 5). shows OTR in the orbital shaking reactor as a function of shaking frequency, filling volume, organic phase loading (circular and triangular markers) as well as with/without the presence of a helical track (black and blue marker colours respectively). Error bars reflect the standard error of the mean at the 95% confidence interval (n = 5). shows productivity results from the bioreactors. Biotransformation was initiated 12 h after induction which in turn took place 12 h after inoculation. Error bars reflect the standard error of the mean at the 95% confidence interval (n = 3).

DETAILED DESCRIPTION OF THE INVENTION

The current invention provides a method of whole-cell catalysed biotransformation of linear alkanes to oxygenated products in actively dividing cells capable of expressing a biocatalyst that catalyses conversion of a linear alkane into an oxygenated product, comprising incubating the actively dividing cells in a biotransformation medium comprising a linear alkane, thereby catalysing the conversion of the linear alkane into the oxygenated product.

In particular, the invention relates to a method of whole-cell catalysed biotransformation of linear alkanes to oxygenated products in actively dividing cells capable of expressing a biocatalyst that catalyses conversion of a linear alkane into an oxygenated product, comprising (i) incubating actively dividing cells in a growth medium including (a) a polysaccharide and (b) a polysaccharide-hydrolysing enzyme that hydrolyses the polysaccharide into a growth substrate for the actively dividing cells at a controlled rate and (ii) incubating the actively dividing cells in a biotransformation medium comprising a linear alkane, thereby catalysing the conversion of the linear alkane into the oxygenated product.

However, it is to be appreciated that, although not a preferred embodiment of the invention, a person skilled in the art may, rather than by use of a polysaccharide and polysaccharide- hydrolysing enzyme to control the rate of growth of the actively dividing cells, use a fed- batch system where a substrate for growth of the actively dividing cells is fed into the system at a controlled rate, thereby to control the rate of growth of the actively dividing cells during the method.

As used herein, the term "growing cell" refers to a cell where growth is supported by a medium that supports active cell division, whereas the term "resting cell" refers to a cell in a medium that does not support active cell division, although the cell may still be metabolically active.

As used herein "bioreactor" refers to a system in which a biological conversion is effected. Accordingly, a bioreactor may range from a small-scale system such as a vial or micro-titre plate to a mid-scale bioreactor such as a flask, to a large fermentation chamber capable of holding many cubic meters of fluid.

Although in a particular embodiment of the invention, the linear alkane is converted into a primary alcohol, it is to be appreciated that the method of the invention may be used for conversion of linear alkanes into a variety of commercially useful oxygenated products, including alcohols, ketones, aldehydes, hydroxyacids, dicarboxylic acids, and the like known to those skilled in the art.

Recently an enzyme-controlled glucose delivery system that mimics a fed-batch mode of operation for enhanced biomass and recombinant protein production, the EnBase ^® technology, (Panula-Perala et al., 2008, Krause et al., 2010, Siurkus and Neubauer, 2011) was developed. The method allows the continuous supply of a carbon and energy source (e.g. glucose) from the polysaccharide substrate through the activity of starch-degrading enzyme. The technology has been applied in bacteria (Siurkus and Neubauer, 201 1 , Glazyrina et al., 2010) and yeast systems (Hortsch and Weuster-Botz, 2011) on a cell unit basis.

The applicants have investigated whether this technology might also be applicable to the field of biotransformation for use during the biotransformation reaction in coversion of linear alkanes to high value oxygenated products, in particular, with the use of growing (actively dividing) cells, rather than resting cells as traditionally used. In addition, the applicants have determined whether such technology could be used during the biotransformation phase of the biocatalytic reaction involving an organic phase. Furthermore, the applicants have investigated whether the technology may be useful to facilitate the use of a single bioreactor, or equivalent small scale system such as a vial, microtitre plate or flask system for the complete duration of the two-phase biotransformation process.

In one embodiment, the method comprises the steps of:

(i) providing cells capable of expressing a biocatalyst that catalyses conversion of a linear alkane into an oxygenated product;

(ii) introducing the cells of (i) into a growth medium that supports active cell division including (a) a polysaccharide and (b) a polysaccharide-hydrolysing enzyme wherein the enzyme hydrolyses the polysaccharide into a substrate for growth of the actively dividing cells at a controlled rate and maintaining active cell division at a controlled rate in the growth medium;

(iii) expressing the biocatalyst in the actively dividing cells of (ii);

(iv) adding a biotransformation medium comprising the linear alkane to the actively dividing cells of (iii);

(v) incubating the actively dividing cells of (iv), thereby catalysing the linear alkane into an oxygenated product; and

(vi) recovering the oxygenated product of (v).

Preferably, at least steps (ii) to (vi) are performed in a single bioreactor.

The step (ii) may comprise a step of inoculating cells grown in an initial growth medium which may either be a conventional growth medium or a medium including (a) a polysaccharide and (b) a polysaccharide-hydrolysing enzyme wherein the enzyme hydrolyses the polysaccharide into a substrate for growth of the actively dividing cells at a controlled rate into the growth medium of (ii). Alternatively, the step (ii) may comprise a step of supplementing the cells grown in the initial growth medium with the growth medium components of (ii). Generally the initial growth medium will differ from the growth medium of (ii) in that although they may both comprise a polysaccharide-hydrolysing enzyme wherein the enzyme hydrolyses a polysaccharide into a substrate for growth of the actively dividing cells at a controlled rate, the concentration of the polysaccharide-hydrolysing enzyme in the growth medium of (ii) is different to that of the first medium.

Additionally, the growth medium of (ii) may further comprise a growth-limiting nutrient and/or at least one inducing agent for enzyme expression. Typically, the inducing agent may be IPTG or lactose, although others may be used that are known to those skilled in the art and may be selected depending on the host organism used for biotransformation.

Typically, the polysaccharide is a starch that can be digested by an enzyme into a readily assimilable substrate by the cells. For example, the substrate could be amylose, amylopectin, dextrin, cellulose, hemicellulose and derivatives thereof known to those skilled in the art.

The growth substrate selected would typically depend on the cell to be cultured and the growth substrate for that cell. For example, the growth substrate could be sucrose, fructose, glucose or another readily assimilable compound hydrolysed from the polysaccharide including maltose, or maltotriose as desired for growth.

The polysaccharide-hydrolysing enzyme may be a single enzyme or a cocktail of enzymes. Typically, the enzyme would be selected depending on the polysaccharide used and the growth substrate desired to be hydrolysed from the polysaccaride. For example, the enzyme may be an amylase, a glucoamylase, an isoamylase, a beta-glucosidase, a cellulolytic enzyme, or others known to those skilled in the art. Preferably the enzyme is glucoamylase. Typically the glucoamylase concentration during the biotransformation stage of step (iv) is in the range of 0.6 to 6 units per liter (U L ¹ ). Optimally, the glucoamylase concentration is 0.6 U L ^"1 .

The medium comprising the polysaccharide and enzyme would typically be a mineral salt medium (MSM). In a particular embodiment, the medium may comprise 2 g/L Na ₂ S0 ₄ , 2.68 g/L (NH ₄ ) ₂ S0 ₄ , 0.5 g/L NH ₄ CI, 14.6 g/L K ₂ HP0 ₄ , 3.6 g/L NaH ₂ PO ₄ .H ₂ 0, 1.0 g/L (NH ₄ ) ₂ -H- citrate and 1.5 M MgS0 ₄ . Optionally, low amounts of tryptone (0.24 g/L) and yeast extract (0.48 g/L) may be added (complex additives). In a preferred embodiment of the invention, the medium of step (ii) is Enbase ^® Flo medium. Optionally, tryptone, peptone and yeast extract may be added as a booster to the medium at step (ii). In one embodiment, the base MSM is supplemented with 3mM MgSCU, 2 ml/L of trace element solution and 0.1 g/l thiamine hydrochloride. In an alternative embodiment, the MSM is supplemented with MOPS-buffer at pH 7.

In an alternative embodiment of the invention, at step (iv), prior to adding the biotransformation medium, the cells are harvested, washed and resuspended in an MSM comprising a metabolically inactive substrate and enzyme but without complex additives, such as peptone. In a specific embodiment of the invention, the medium used is the Enbase ^® Flo Zero medium (obtained from BioSilta Oy, Oulu, Finland). However, preferably, rather than harvesting, washing and resuspending the cells, the medium prior to step (iv) is simply supplemented to obtain the MSM comprising the metabolically inactive substrate and enzyme, without complex additives. In this method, use of a single bioreactor or similar small-scale system is facilitated.

In a preferred embodiment of the invention where the cell to be used in the biotransformation method is Escherichia coli, the polysaccharide and is digested by the polysaccharide hydrolysing enzyme, glucoamylase (or EnZ 'lm), thereby releasing the growth substrate, glucose.

The method may be used for any linear alkane and the biocatalyst would be selected depending on the alkane to be converted and the desired catalytic reaction. In particular, any n-alkane from a Ci - C36 alkane may be used.

Various biocatalysts may be used, although typically the biocatalyst is an oxygenase. Preferably, the biocatalyst is a hydroxylase. In particular, the biocatalyst may be AlkB or a cytochrome P450 like CYP153, CYP102 or CYP52. Most preferably, the cytochrome P450 is CYP153A6.

In a particular embodiment of the invention, the n-alkane is octane and the octane is catalysed by the cytochrome P450 CYP153A6 into 1 -octanol.

The method may be used with various cells capable of growth in cell culture. The method would optimally be used with a bacterial or a fungal cell, particularly a yeast cell. The cell selected for use may depend on the alkane feedstock to be converted and the biocatalytic enzyme(s) intrinsically expressed by the cell.

Various such cells are well known to those skilled in the art. For example, the bacterial cell may be Pseudomonas sp. , including Pseudomonas putida, Rhodococcus sp. , Acinetobacter sp., Bacillus sp., including Bacillus megaterium, Mycobacterium sp, Arthrobacter sp., Streptomyces sp., Marinobacter aquaelolei, or Geobacillus thermodenitrificans.

The fungal cell may be a yeast cell including Candida sp., Yarrowia or Arxula.

In a preferred embodiment of the invention, the cell does not intrinsically express a biocatalytic enzyme, but has been genetically engineered to express a desired biocatalyst. Typically, the genetically engineered cell is Escherichia coli (E. coli). Various methods of generating recombinant E. coli cells to express heterologous proteins including biocatalyst enzymes are well known to those skilled in the art.

The cell may further express or be genetically engineered to express one or more electron transfer protein(s) for providing reducing equivalents for biotransformation. It is to be appreciated that the electron transfer protein selected would depend on the orgin of the chosen P450 biocatalyst. For example, the electron transfer protein may be any one or more of Ferrodoxin, Ferrodoxin Reductase, Rubredoxin, or Rubredoxin Reductase, Cytochrome P450 Reductase, or FAD/FMN Reductase.

In an alternative embodiment of the invention, the cell may further express or be genetically engineered to express a transport protein for transporting hydrophobic substrates or products or both across the cell membrane for recovery. One example of a transport protein that may be used is AlkL, although others known to those skilled in the art may be selected.

Similarly, the cell may further express or be genetically engineered to express or over- express a dehydrogenase enzyme, including a glucose dehydrogenase or glycerol dehydrogenase enzyme or others known to those skilled in the art to produce an overexpression of reducing power.

Although the cells may constitutively express the biocatalyst, it is more typical that the method at step (iii) would comprise a method of inducing the cells to express the biocatalyst. In a preferred embodiment of the invention, the cells would be induced during their mid- exponential growth phase (A578 of about 1.2 - 1.5). Various inducers may be used known to those skilled in the art. For example, the cell may be an E. coli cell expressing β- galactosidase, and the induction may be by addition of IPTG or lactose. Further, in a preferred embodiment of the invention, the cell medium is supplemented at induction with 0.25 mM δ-aminolevulinic acid (δ-ALA) and 50 μΜ FeCI ₃ .6H ₂ 0.

The biotransformation medium may further comprise an organic solvent, thereby to produce a product sink, wherein during step (vi) the oxygenated product is drawn into the product sink for recovery. It is to be appreciated that the product sink may be selected by those skilled in the art depending on the oxygenated product to be recovered, the toxicity level of the solvent sink and the effect of the solvent sink on the particular organism used. It is further to be appreciated that the linear alkane itself may serve as the organic solvent during two-phase biotransformation.

For example, where the product to be recovered in an alcohol such as octanol, the solvent may be 1-hexadecanol or bis(2-ethylhexyl) phthalate (BEHP). Preferably, the solvent is BEHP.

Step (v) may be performed at from about 20 to about 37 °C. Optimally however, step (v) is performed at from about 20 °C to about 25 °C.

The method is versatile in that it may be performed on a small scale such as with the use of vials or microtitre plates and the like, or the method may be scaled up for use in a bioreactor.

For example, at least from steps (ii) onward of the method can be performed in a bioreactor such as a stirred tank reactor or an orbital shaking reactor, or others known to those skilled in the art. Alternatively, the entire method may be performed in a bioreactor. Preferably, the orbital shaking reactor would comprise a helical track with dimensions selected according to standard methods known to those skilled in the art.

It is to be appreciated that conditions including the type of bioreactor, agitation rate, aeration rate and the like may be optimised depending on the type of bioreactor selected, and the desired size and capacity of the reactor according to standard methods known to those skilled in the art.

The invention will be described by way of the following examples which are not to be construed as limiting in any way the scope of the invention.

EXAMPLE 1

1. Small-scale biotransformation with whole-cell biocatalysts

1.1. Experimental approach

As indicated in Table 1 , the first sets of biotransformation experiments were conducted using E.coli cells cultured on glucose-based chemically defined medium. Biotransformations carried out with growing whole-cell biocatalysts were compared to resting cell-catalysed bioconversion, both operated as a batch process (Process A). This allowed a view into the possible limitations of using a growing biocatalyst system under biotransformation conditions in a batch reaction. The subsequent experiments were performed using the EnBase® technology. Optimised cultivation and induction conditions were established for maximum production of P450. Resting cells originating from the experiment were applied in octane hydroxylation reactions to ensure comparable performance with cells grown on LB and the glucose-based chemically defined medium (Process B). Subsequent experiments were conducted using growing E.coli cells, expressing P450, in EnBase. As recommended in the BioSilta protocol for proper protein expression and pH effects, the initial experimental studies were conducted with addition of complex boosting reagents (Process C). However, the addition of complex additives presented an undefined medium which prevented process control. Therefore, further experiments were conducted with EnBase® medium to which complex reagents were not added prior to start of bioconversion (Process D). Alternatively, cells were harvested from EnBase® growth medium and placed in fresh reaction media devoid of complex reagents for the bioconversion stage (Process E). Here, the glucose release rate was varied by using different gluco-amylase concentrations within the bioreaction mixture. The octanol concentration was quantified under these conditions and related to cell metabolism during biotransformation. The summary is presented in Table 1.

Table 1 : Overview of experimental approach using growing cells as biocatalyst in the terminal hydroxylation of octane to 1 -octanol.

Process Cultivation Medium Biotransformation Conditions Mode of catalyst

Prior to Start of

Bioconversion

A Glucose-based Glucose based defined media - Batch (a) Growing and (b) Resting chemically defined process whole-cells

B EnBase® (I) 0.2 M Phosphate buffer (7.2) Resting-whole cells

containing glucose - Batch process

(II) Cells resuspended in starch solution,

supplemented with glucoamylase

C EnBase® Octane was added to growing cells on Growing-cells

EnBase® medium. At the same time,

booster containing complex additives

was added as recommended by

BioSilta. Fed-batch process

D EnBase® Octane was added directly to growing Growing-cells

cells in EnBase® medium after 12 h,

with no inclusion of booster. Fed-batch

process

E Enbase® Cultivated and induced cells were Growing cells

harvested and resuspended in EnBase®

Flo Zero (no booster additives).

Thereafter, octane was added to resume

biotransformation. 1.2. Methodology

1.2. 1. Medium composition

The EnBase® medium is a minimal salt medium containing a starch polysaccharide as substrate for glucose release, vitamins, trace elements and complex nutrient components, referred to as booster (Krause et al., 2010). In the protocol for its use, the addition of a glucoamylase, the polysaccharide-degrading enzyme (EnZ 'Im) responsible for the release of glucose, is specified. A variety of the EnBase® preparation, EnBase® Flo Zero, is devoid of complex nutrients (e.g. peptone). The complete EnBase® media packs (Flo and Flo Zero) were purchased from BioSilta Oy (Oulu, Finland). For cultivations in baffled shake flasks, 0.1 ml L ^"1 antifoam was added to prevent foaming and all media contained 30 μg ml ^-1 kanamycin for maintenance of plasmid stability through the extended growing phase.

12.2. Precultures

The seed cultures were prepared by inoculating LB broth supplemented with kanamycin (30 μg ml ^-1 ) with E. coli cells from glycerol stocks previously maintained at -60°C. This was incubated on an orbital shaker at 37°C for 6 to 8 h at an agitation speed of 200 rpm.

1.2.3. Process A: Octane biotransformation using resting and growing cells

cultivated on chemically-defined glucose medium

A 4 ml inoculum was used to inoculate the glucose-based defined medium (200 ml in 2 L flask, 2% v/v). Selective pressure was maintained by adding 30 μg ml ^"1 kanamycin and the culture was incubated at 30°C with constant shaking at 160 rpm. At mid-exponential phase of cell growth (A578 of 1.2 - 1.5), the medium was supplemented with 0.25 mM δ- aminolevulinic acid (δ-ALA), 50 μΜ FeCl3.6H ₂ 0 and 0.25 mM isopropyl^-D- thiogalactopyranoside (IPTG). Cultures were further incubated at 20°C, shaking at 160 rpm for a total of 24 h after inoculation. Thereafter, the cells were harvested through centrifugation (13 000 g for 10 mins, 4°C) and used for octane biotransformations as both metabolically active resting cells and growing whole cell biocatalysts. The resting cells were resuspended at up to 3.0 gpcw LBRM ^"1 in 200 mM sodium phosphate buffer (pH 7.2), devoid of a nitrogen source. For the growing biocatalyst system, the cells were resuspended in chemically-defined glucose medium to a biomass concentration of 2.5 - 3 gpcw LBRM ^"1 . One millilitre aliquots of this mixture were transferred into 60 ml sterile, amber screw-cap vials and 200 μΙ of n-octane and 100 μΙ of BEHP were added to each vial before capping and placing on an orbital shaker at 200 rpm and 20°C. The reaction vials were opened occasionally (6 to 8 hourly) for the addition of glucose and exchange of gases to prevent carbon and oxygen limitation. Vials were removed at specified time intervals for product extraction, where the full contents of one vial represented a sample. The biotransformation reactions were stopped by adding 100 μΙ of 5 M HCI to each vial. The octanol, residual glucose and acetic acid concentrations were determined. The pH of the reaction mixture was followed offline with a pH electrode (Cyber Scan 2500, Eutech Instruments).

12.4. Process B: Octane biotransformation using resting cells cultivated on

EnBase® medium

A 1 ml aliquot of the preculture was used to inoculate 50 ml of the EnBase ^® Flo medium (2%, v/v) containing the complex polysaccharide which had been introduced into a sterile 500 ml baffled flask. The starch-degrading enzyme (EnZ I'm) was added to the medium at a concentration of 0.6 U L ^"1 . An AirOtop membrane (Thomson Instrument Company, USA) was used to cover the baffled flask for improved air inflow and enhanced oxygen transfer. The culture was incubated on an orbital shaker at 30°C and 200 rpm. At specified time intervals (8, 17 or 22 h) representing different phases of cell growth (As78nm of 2.6. 7.2 and 10.4 respectively), the broth was supplemented with 0.5 mM IPTG (for induction of protein expression) and 0.5 mM δ-ALA (for heme synthesis). The booster component of the EnBase ^® medium was also added for optimal pH conditions (Yun et ai, 1996) and an additional 0.6 U L ¹ of glucoamylase was included. This cultivation was allowed to continue for a total duration of 48 h and the P450 concentration determined. The cells were harvested through centrifugation (13 000 g for 10 mins at 4°C), washed and resuspended in 200 mM sodium phosphate buffer (pH 7.2) to a concentration range of 3.5 - 4 gpcw LBRM ^"1 for octane biotransformation as resting whole-cell biocatalysts. The bioreaction mixture (1 ml) in 60 ml reaction vials was supplemented with glucose (5 g L ¹ ) as the carbon and energy source.

In another set of experiments, the biomass was resuspended in a 30 g L ¹ starch solution and supplemented with glucoamylase (0.6 U L ¹ ).

For both studies, the organic layer consisting of 200 μΙ octane (substrate) and 100 μΙ BEHP (as product sink) was added to initiate whole-cell biocatalysis. The vials were incubated on an orbital shaker at 200 rpm and 20°C, and were removed at specified time intervals across the range of 3 to 120 hours for product extraction. The biotransformation reactions were stopped by adding 100 μΙ of 5 M HCI to each vial. 12.5. Process C: Octane biotransformation using growing cells cultivated on EnBase® medium including complex additives

Following optimisation of conditions for maximum P450 expression and activity, biotransformation of octane was studied using growing whole cell biocatalysts. The cells were cultivated in EnBase® medium. After the addition of IPTG (0.5 mM) , δ-ALA (0.5 mM), EnZ I'm (0.6 U L ¹ ) and the booster component, the E. coli culture was incubated for 3 h at 20°C. Thereafter, a 1 ml portion of the broth was pipetted into each 60 ml reaction vial and the organic phase consisting of octane (200 μΙ) and BEHP (100 μΙ) was added. The reaction vials were agitated on an orbital shaker at 200 rpm under varied temperature conditions (20, 25 and 30°C). Vials were removed at specified time intervals across the range 3 to 120 hours for product extraction. The biotransformation reactions were stopped by adding 100 μΙ 5 M HCI to each vial and the octanol concentration was quantified using GC analysis.

12.6. Octane biotransformation using growing cells on EnBase® medium without complex additives

In the subsequent studies using a growing cell system in EnBase ^® medium, the complex additives were excluded from the bioconversion phase. This was necessary for a well defined system under varied EnZ I'm concentrations to achieve different glucose release rates. This informs the relationship/s between cell metabolism, glucose uptake rate and product formation. The EnBase ^® Flo Zero medium (BioSilta Oy, Oulu, Finland), i.e. without complex additives, was used in this study.

12.6. 1. Glucose release kinetics and cell growth

A 1 ml portion of EnBase ^® medium consisting of the complex polysaccharide was introduced into sterile 60 ml reaction vial. The glucoamylase (EnZ I'm) concentration was varied from 0.6 to 12 U L ^"1 and incubated at 20°C. For another batch of reactions, 200 μΙ of octane was added to the vial to investigate the possible inhibiting effect of the substrate on EnZ I'm. Samples were taken at specified intervals and the reaction terminated by the addition of 50 μΙ 5M HCI. Glucose formation was analysed by determining the generation of H2O2 spectrophotometrically, following reaction catalysed by glucose oxidase using a glucose oxidase kit (Roche, Germany).

In a follow-up study, cell growth on EnBase® medium was investigated in 60 ml vials under varied glucoamylase activities (0.6 to 12 U L ¹ ). The EnBase® medium (50 ml) was inoculated with the seed culture (2% v/v). A 1 ml aliquot of the well mixed broth was immediately transferred to each vial and the specified concentration of glucoamylase added. The capped vial was incubated at 20°C and agitated at 200 rpm. The reaction vial was periodically (6-8 hourly) opened to air under sterile conditions to avoid oxygen limitation. Samples were taken at specified time intervals by sacrificing a vial set. Cell growth was monitored through absorbance reading at As^ and the pH of the medium was followed offline with a pH electrode (Cyber Scan 2500, Eutech Instruments).

1.2.6.2. Process D and E: Octane bioconversion using growing biocatalysts A 1 ml aliquot of the preculture was used to inoculate 50 ml of the EnBase ^® Flo medium (2%, v/v) containing the complex polysaccharide which had been introduced into a sterile 500 ml baffled flask. Glucoamylase (0.6 U L ¹ ) was added to the culture and an AirOtop membrane was used to cover the baffled flask. The culture was incubated on an orbital shaker at 30°C and 200 rpm.

In an experimental set, the culture was induced with 0.5 mM IPTG (for protein expression) and 0.5 mM δ-ALA (for heme synthesis) after 12 - 14 h (As78nm of 6.2). The culture was further incubated for 3 h (at 20°C) in preparation for the biotransformations (Process D). A 1 ml portion of the cell broth (6 g Dew L ^"1 ) was pipetted into each 60 ml vial under varying concentrations of EnZ I'm (0.6 to 12 U L ¹ ). The biotransformation was initiated with the addition of octane (200 μΙ) and BEHP (100 μΙ).

In another set of experiments, the culture was induced with the same concentration of IPTG and δ-ALA at the 24 ^th hour (A ₅₇ 8nm of 16.5) and further incubated for 3 h at 20°C. The biomass was harvested through centrifugation (7000 rpm for 10 mins at 4°C), washed with sterile distilled water and resuspended in EnBase ^® Flo Zero to a concentration of 1 to 1.5 g L ^"1 (Process E). The broth was supplemented with 0.5 mM IPTG and δ-ALA including 100 μΙ of thiamine solution included in the BioSilta pack. A 1 ml aliquot of the bioreaction mixture was pipetted into 60 ml reaction vials and EnZ I'm was added in the concentration range 0.6 to 12 U L ¹ . This was followed by the addition of octane (200 μΙ) and BEHP (100 μΙ).

For both studies, the reaction vials were agitated on an orbital shaker at 200 rpm and 20°C. Vials were removed over the duration of 120 hours (where the full contents of one vial represented a single sample) and the reaction terminated by adding 100 μΙ 5 M HCI prior to product extraction and quantification. The pH of the reaction mixture was measured offline with a pH electrode (Cyber Scan 2500, Eutech Instruments).

The summary of the experimental set-up for each reaction set is detailed in Table 2. 1.3. Results

1.3. 1. Process A: Octane biotransformation using resting and growing whole cell biocatalyst on glucose based defined medium

Cells cultured on glucose based chemically defined medium were used for octane bioconversion as a resting metabolically active biocatalyst. The result was compared with growing cells under same biotransformation conditions. The octanol production, P450 concentration, acetic acid formation and reaction pH are presented in Figure 1. The maximum octanol formation rate using the growing whole cell biocatalysts in glucose-based defined medium was 0.20 g LBRM ^"1 IT ¹ . This was achieved at a glucose uptake rate of 0.68 g L ^"1 h ^"1 over the initial duration (8 h) of octane biotransformation and represents a two fold increase over the rate in resting cells. This was apparently responsible for the 2.5 fold increase in octanol formation rate using growing whole-cells as biocatalyst. However, the maximum product formation rate in the growing system was only maintained for 25 h, leading to a final octanol concentration of 4.35 g LBRM ^"1 (0.91 goctanoi gDcw ^" ) in 140 h (Figure 1a).

CO

(Z

on

CO m

CO

I

m

m

H

3J

(Z

I- m

CD

NB: In all biotransformation studies, 200 μΙ octane and 100 μΙ of BEHP were added.

Biomass concentration was increased from 2.5 to 4.8 g L ¹ at the end of biotransformation. A similar final volumetric concentration (4.5 g LBRM ^"1 ) was observed at 140 h with resting whole cells, although with reduced maximum volumetric productivity and lower biomass concentration. This resulted in higher octanol yield on biocatalyst,

1.44 goctanol gDCW ^" ■

The P450 concentration profile in Figure 1 b shows that growing biocatalysts were able to synthesise the heterologous CYP153A6 under biotransformation conditions. While the active properly folded P450 concentration was depleting under resting whole cell biocatalysis, it was increasing in the growing system up to 25 h. The reduced efficiency of the growing whole cell biocatalyst after 25 h may likely be due to acetic acid accumulation (Figure 1c) and the resulting acidification of the reaction mixture (Figure 1d). The maximum volumetric production rate of acetic acid was 0.08 and 0.02 g LBRM ^"1 IT ¹ in growing and resting whole cell biotransformation respectively. The final acetic acid concentration (5 g L ¹ ) reached with the application of growing biocatalyst in 72 h was twice the value achieved with the resting biocatalyst system over the same duration. This led to a drop in pH (to 4.6) and fast acidification of the reaction mixture (pH less than 6 in 18 h) which apparently resulted in the inhibition of whole cell biocatalyst. These results show the potential of growing biocatalyst system; however the accumulation of acetate from increased cell metabolism requires resolution, possibly through the introduction of a fed-batch process. The intermittent addition of glucose represents a "pseudo" fed-batch system but does not present the nutrient limited operation typical of a fed-batch process.

1.3.2. EnBase® Technology

1.3.2. 1. Characterisation of the glucose delivery system

The EnBase ^® delivery system is based on the controlled glucose release into the biotransformation mixture through enzymatic degradation of the starch polymer. The concentration of the polysaccharide degrading enzyme glucoamylase was varied to vary the rate of glucose release. This, in turn, informs the optimum concentration that favours cell growth and octane oxidation. In the initial experiments, the impact of octane on glucoamylase activity was studied. Figure 2 shows that glucose accumulation and the associated glucose release rate are a function of the glucoamylase concentration over the range, 0.6 to 12 U L ^"1 . The results also show that presence of octane at 30%, v/v did not significantly affect glucoamylase activity. No significant difference (P< 0.05) in glucose release rate was noticed comparing the control (no octane) and experimental samples (i.e. addition of octane as depicted in biotransformation conditions). This is represented in Figure 2a through the low standard deviation achieved on averaging data collected in the presence and absence of octane. At lower enzyme levels, glucose was released continually into the system over the entire duration of 80 h, thereby showing that glucoamylase activity can be sustained over a long duration.

The initial glucose release rate (from Figure 2b) over the first 10 h shows a linear relationship with the corresponding increase in gluco-amylase concentration (Figure 3). It is expected that this correlation will be maintained provided glucose accumulation (that prompts feedback inhibition) or overflow metabolism is avoided during cell cultivation or biotransformation. This result indicates that the gluco-amylase can function in the two phase biocatalytic system involving aqueous and organic phase.

1.3.2.2. Process B: Octane bioconversion using resting whole cell E. coli

cultured on EnBase® medium

Further experiments were conducted with EnBase ^® medium, a technology that allows the supply of glucose in a controlled release fashion as is typical of the controlled feed in a fed- batch process. Initial studies were performed using resting cells grown on EnBase ^® medium and compared with the whole cell activities reported for cells grown on LB and glucose based chemically defined medium (data not shown). Optimum conditions for maximum gene expression were established through induction studies. It was observed that the addition of IPTG and amino-levulinic acid (heme precursor) at 8, 17 and 22 h did not negatively impact the final biomass concentration from cell cultivation (Table 3). The culture reached a final biomass concentration in the range of 10.5 to 12 g Dew L ^"1 . The bioconversion was carried out in phosphate buffer with glucose addition. Results presented in Table 3 show that protein induction at the 22 h gave the best result in terms of P450 expression, whole-cell biocatalyst activity, biocatalyst efficiency and volumetric octanol concentration. The maximum biocatalyst activity achieved at 20 and 37°C show a 25% increase over the maximum value previously reported with the use of LB medium under the same conditions (data not shown). A final octanol yield of 1.93 g gpcw ^"1 was observed after 94 h of biotransformation, representing a 1.5 to 2 fold increase over the concentration achieved with resting biocatalyst expressing CYP153A6 cultured on LB medium (data not shown). The pH at the end of reaction (94 h) was less than 6 in all the reactions, suggesting release of acidic metabolites or by-product into the reaction mixture. Table 3: Growth and induction studies on EnBase® media over 45 h to optimise P450 production while also maximising biocatalyst activity for octane hydroxylation.

Time of Final P450 Biocatalyst Activity Biocatalyst Maximum induction Biomass (μιηοΙ g _D cw ^"1 ) μιηοΙ (gDcw min) ¹ Efficiency Octanol (h) (gDcw L ¹ ) 37°C 20°C (g gDcw ¹ ) (g LBR ¹ )

20°C, 94 h

8 11.85 0.06 2.65 0.45 0.18 0.78

17 10.45 0.06 16.28 4.99 1.39 5.1

22 11.10 0.15 20.37 6.28 1.93 7.5

*** Biomass grown on EnBase ^® was washed and resuspended in sodium phosphate buffer supplemented with glucose. Biomass concentration was in the range 3.7 - 4.3 g LBRM ^"1 . Biotransformation were carried out in 1 ml volumes in 60 ml vials.

In another set of experiments, the biomass from the optimised induction condition (i.e. protein induced at 22 h) was resuspended in starch solution, supplemented with glucoamylase across the activity range 0.6 - 12 U L ¹ and used for octane bioconversion. The continuous supply of glucose was guaranteed through glucoamylase activity. The biocatalyst activity across all glucoamylase concentrations proceeded at an initial and maximum rate of 0.20 g octanol formed L ¹ h ^"1 (3.95 μηιοΙ gDcw ^"1 min ^-1 ). The maximum octanol concentration, an average of 8.67 ± 0.10 g LBRM ^"1 , was achieved in 95 h using a glucoamylase concentration in the range 0.6 to 3.0 U L ^"1 (Figure 4a). With increase in glucoamylase activity (6 and 12 U L ^"1 ), a 20% drop in octanol concentration was observed. This was apparently due to the acidification of the medium under biotransformations conditions (Figure 4b). In systems with a reduced glucose release rate, the pH of the medium was relatively stable in the range 6.4 to 7 for most of the reaction period (60 h), only reaching its lowest point of 6.2 at 95 h.

1.3.2.3. Process C: Octane oxidation by the growing whole cell biocatalyst cultured on EnBase® medium including complex additives

The hydroxylation of n-octane was carried out by growing E. coli cells cultured on EnBase ^® medium following induction under the optimal conditions determined previously (i.e. induction of protein expression at 22 h of cell growth). The biotransformation study was carried out under temperatures of 20, 25 and 30°C. The results are presented in Figure 5 and compared with standard biotransformation condition using growing biocatalyst in defined medium at 20°C, described above. As with previous studies, the reaction at lower temperature (20°C) gave the highest biocatalyst efficiency compared to higher temperature conditions at 25 and 30°C. The reaction proceeded at an initial rate of 0.30 g L ¹ h ^"1 (5.52 μηιοΙ gpcw ^"1 min ^-1 ) reaching a final octanol concentration of 15.5 g LBRM ^"1 (~ 2.2 g gDcw ^"1 ) in 96 h. This significant increase in octanol concentration over previous results (data not shown) in terms of volumetric and biocatalyst efficiency shows an improved hydroxylation reaction with the growing cell system in the EnBase ^® medium. This was likely due to the influence of booster tablet containing complex additives, yeast extract and tryptone. This supports the process with required amino acids, cofactors and vitamins and also stabilises the pH of the medium. To have a better understanding on the influence of slowly released glucose on cell metabolism and octanol production, further biotransformation experiments were carried out with EnBase ^® medium lacking the booster component containing complex additives. In addition, the rate of glucose release was varied.

1.3.2.4. The influence of glucose supply rate on cell growth in EnBase®

medium

To encourage process control, further biotransformation studies using growing cells were conducted with EnBase® medium lacking the booster component consisting of complex additives. Studies were initially conducted to determine the effect of glucose release rate on cell growth rate and biomass accumulation. This was done by varying glucoamylase concentration from 0.6 to 12 U L ^"1 . The results shown in Figure 6 demonstrate that cell growth rate can be manipulated under varied glucoamylase conditions, confirming that cell growth is glucose limited. A two fold increase in biomass concentration was observed at enzyme concentration of 12 U L ¹ compared to that achieved at 0.6 U L ^"1 . The maximum growth rate, observed at an amylase activity of 6 - 12 U L ¹ , represented a 10 fold increase over that at 0.6 U L ^"1 . The overview of the results, presented in Table 4, suggested overflow metabolism at higher glucose release rate, indicated by the decreasing pH. This was more likely than anaerobic metabolism as the vials were consistently opened under sterile conditions to allow exchange of air, thereby discouraging oxygen limitation. The resultant production of acidic metabolites and associated reduced pH may become inhibitory to the cells. Table 4: Key parameters describing cell growth under varied glucoamylase concentration over a period of 48 h. The reaction was carried out in 60 ml vial using 1 ml of cell broth (2% v/v inoculum), and incubated at 20°C.

Enzyme Level Final A578nm Final Umax PH

(U L ¹ ) Biomass (h- ¹ )

(g L- ¹ )

0.60 7.72 3.75 0.02 6.60

1.20 9.42 4.87 0.03 6.43

3.00 12.46 7.37 0.13 6.13

6.00 17.12 8.25 0.21 5.40

12.00 16.50 8.87 0.21 4.20

1.3.2.5. Process D: Octane bioconversion using the growing whole-cell

biocatalyst cultured on EnBase® without addition of complex organics in the biotransformation phase

Following the knowledge acquired from glucose release rate at differing enzyme concentration, the EnBase ^® medium was applied in octane bioconversion process in the absence of complex additives. The octanol accumulation over the initial phase of 9 h is presented in Figure 7 as a function of glucose release rate. This does not follow a linear trend similar to the glucose release profile at the maximum glucoamylase concentration. The octanol concentration did not correlate to the glucose release rate over the 9 h duration. This suggests that glucose released from starch degradation at high release rates was also channelled towards other pathways and functions within the cell. No acetate was detected at lower amylase concentrations. At 12 U L ¹ , 0.7 g L ¹ acetic acid was observed.

Over a longer duration (60 h), it was observed that the reaction with the maximum concentration of starch degrading enzyme (12 U L ¹ ) produced less octanol compared to processes with lower concentration of glucoamylase (0.6 - 6 U L ¹ ) (Figure 8a). The final octanol concentration was 30% less than observed at lower glucoamylase concentrations (0.6 - 6 U L ^"1 ). The final octanol achieved was almost the same in systems operated with 0.6 to 3 U L ^"1 and slightly lower with 6 U L ¹ amylase system. The trend of acetic acid formation was consistently higher at the maximum glucoamylase activity, reaching a final acetate concentration of 3.8 g L ¹ (Figure 8b).

The rapid and increased acetate production in the presence of 12 U L ¹ led to a severe drop in pH, below 6, as early as the 20 ^th hour becoming toxic to the biocatalyst. Octanol was not produced thereafter. In the systems with lower glucoamylase concentration, overflow metabolism was reduced as suggested by the lower concentration of acetic acid. The final concentration was in the range of 2.8 to 3.2 g L ^"1 . This suggests that where glucose release rate exceeded the glucose requirement to support bioconversion, resulting in accumulation (informed by the varied concentration of glucoamylase), the excess glucose was channelled towards other cell functions including biomass formation.

As summarised in Table 5, biomass accumulation was highest with the maximum glucose release rate. This increased by 2 fold from starting concentration of 6 g L ¹ while only 50% increase was observed at glucoamylase level of 0.6 U L ¹ . This indicates that a substantial part of the released glucose at high glucoamylase level was used for biomass formation. Octanol formation was slowed down and finally inhibited by acetic acid formation and the resulting medium acidification below pH of 6 (Table 5). This and the usage of "excess" glucose for other cell functions including biomass formation led to reduced biocatalyst efficiency at higher glucoamylase concentration (0.40 g octanol gocw ^"1 ). Specific octanol formation was higher at lower glucoamylase concentration, suggesting that the limited glucose release contributed to the efficient biotransformation of octane to octanol under this condition.

Table 5: Hydroxylation of n-octane using growing whole E. coli (expressing CYP153A6) on

EnBase ^® medium without booster reagent during the bioconversion phase. Biomass concentration at start of biotransformation was 6 gocw LBRM ^"1 while total reaction time was 60 h

Amylase Initial Final Max Final Biocatalyst Final pH at

(U L- ¹ ) Glucose Biomass Octanol Octanol Efficiency Acetic 60 h

Release (g LBRM ¹ ) Production (g LBRM ^"1 ) (goctanol Acid

(g L- ¹ h- ¹ ) rate gocw ^-1 ) (g LBRM ^"1 )

(g L- ¹ h- ¹ )

0.6 0.04 9.0 0.22 6.97 0.77 2.7 5.8

1.2 0.08 10.6 0.25 7.42 0.70 2.9 5.6

3 0.17 1 1 .4 0.27 7.30 0.64 3.2 5.6

6 0.36 12.0 0.27 6.86 0.57 3.3 5.5

12 0.72 12.2 0.27 4.82 0.40 3.7 5.3

*** Summary of data presented in Figure 8. Following 22 h of cell cultivation, then 3 h of induction, 1 ml aliquots were distributed into 60 ml reaction vessel with glucoamylase added in the range 0.6 to 12 U L ^"1 . Thereafter, 200 μΙ of octane and 100 μΙ of BEHP (product sink) were added to commence biotransformation at 20°C and agitation at 200 rpm.

1.3.2.6. Process E: Biotransformation undertaken with resuspended growing cells in EnBase® Flo Zero

In another round of studies, the biomass grown on EnBase ^® medium was harvested and resuspended in EnBase ^® Flo Zero, a medium without complex additives. This was necessary to enable cell resuspension in fresh medium, lacking any excreted metabolite or by-product that may have resulted from cell cultivation. Further, the harvested biomass can be resuspended to lower concentration which was a difficult task to accomplish previously considering that the cells were taken directly from actively growing cells (Table 2). The lower biomass concentration reduced the potential of oxygen limitation. The octanol accumulation trend, biomass accumulation, acetic acid formation, pH profile and residual glucose concentration are presented in Figure 9.

The initial volumetric octanol formation rate (0.1 g L ¹ h ^"1 ) over the first 6 h of reaction was similar across the glucoamylase concentration studied, 0.6 - 12 U L ¹ (Figure 9a). However, a major difference was observed in the final octanol concentration achieved. The maximum concentration of 3.33 g _oc tanoi LBRM ^"1 was achieved over 120 h of bioconversion at glucoamylase concentration of 12 U L ¹ . This was increased by almost 2 fold at lower glucoamylase level (0.6 U L ¹ ). The result presented in Figure 9b suggests that a significant portion of the carbon source was channelled towards biomass formation under conditions of high glucose release rate.

The high glucose release apparently resulted in increased biomass formation over the first 20 h. This was inhibited with increasing duration of bioconversion, finally resulting in a biomass concentration of 5 g L ¹ . This is 60% less than the biomass concentration achieved on EnBase® medium without bioconversion (Table 4). It is also consistent with other levels of glucoamylase concentration, 0.6 to 6 U L ^"1 . This may have been used for octane bioconversion or may be due to energy spilling and carbon sink channelled towards formation of by-products or undesired metabolites.

The acetic acid profile (Figure 9c) suggests that increasing glucoamylase concentration (and thereby initial glucose release rate and potentially residual glucose concentration) encourages overflow metabolism. At 40 h, the reactions with a higher level of glucoamylase produced over 4 g L ¹ of acetic acid which was apparently inhibitory to the whole cell biocatalyst. This corresponded to the time at which octanol production began to level off. Meanwhile, the acetic acid accumulated in the presence of 0.6 U L ¹ glucoamylase remained in the range 1.5 to 1.8 g L ¹ throughout the reaction phase, permitting continued production of octanol. The accumulation of acetic acid reduced the pH of the biotransformation mixture (Figure 9d).

Hence, the increasing concentration of glucoamylase in the reaction mixture led to an inverse effect on pH. The pH decreased to less than pH of 5 in 60 h at the higher glucoamylase concentrations, 3 - 12 U L ¹ . This was confirmed as inhibitory to the system as octanol production levelled off (Figure 9a). The pH was maintained in the range 6.4 - 6.7 at glucoamylase concentrations of 1.2 U L ¹ and less. This shows that overflow metabolism was limited at a reduced glucose release rate, thereby making carbon available for energy generation and cofactor regeneration. This was confirmed in the profile of un-used glucose present in the reaction mixture (Figure 9e). The glucose concentration was maintained in a limited fashion at glucoamylase level below 1.2 U L ¹ and much increased at higher amylase concentrations, 3 to 12 U L ^"1 . This was apparently responsible for overflow metabolism and contributed substantially to the significant level of acetic acid detected at glucoamylase concentration in the range 3 to 12 U L ^"1 . As summarised in Table 6, the reduced carbon wasting and possibly energy spilling contributed to the more efficient bioconversion observed at reduced glucose release rate. Biocatalyst efficiency was 2.1 g octanol gDcw ^" , representing over 2 fold increase over that achieved with growing cells in glucose-based chemically defined medium operated as a batch process.

Table 6: Octane biotransformation using growing cells resuspended in EnBase® Flo Zero, devoid of complex additives. The biomass was grown on EnBase ^® for 22 h, induced with IPTG and incubated at 20°C for 3 h. Harvested biomass was re-suspended (1-1.5 gocw L ^"1 ) in EnBase ^® Flo Zero supplemented with IPTG, δ- ALA and thiamine. A 1 ml aliquot of this mixture was introduced into the 60 ml amber vials, glucoamylase was added at varied concentration followed by octane and BEHP. Total duration was 120 h.

Amylase Final Max Final Biocatalyst Biocatalyst Final pH at

(U L- ¹ ) Biomass Octanol Octanol Activity Efficiency Acetic 120 h

(g LBRM ¹ ) Production (g LBRM ^"1 ) (Mmol g _D cw ^"1 (goctanol Acid rate min ¹ ) gocw ^-1 ) (g LBRM ^"1 )

(g L- ¹ h- ¹ )

0.6 3.00 0.09 6.32 7.20 2.1 1 1 .48 6.40

1 .2 3.00 0.09 5.52 6.90 1 .75 1 .84 5.56

3 3.20 0.1 1 4.49 5.63 1 .40 3.18 5.14

6 4.10 0.12 3.89 5.30 0.95 3.34 4.88

12 5.00 0.13 3.47 4.62 0.67 4.14 4.78

1 .4. Discussion

The use of growing whole cell E. coli expressing CYP153A6 was investigated for improved process productivity in octane hydroxylation. In the analytical scale biotransformations conducted with growing cells, initial studies focused on understanding the process limitations in a batch system. The use of glucose based chemically defined media was favoured over complex LB medium to facilitate process control. Although cells grown on LB medium have been shown to offer higher activity in terminal octane oxidation (data not shown), the undefined nature of the medium discourages its use for biotransformation while applying growing cells as biocatalysts. It will be difficult to achieve the necessary feeding design typical of fed-batch operation as the limiting substrate is unknown.

The overview of the results summarising key data are presented in Table 7. In the study with glucose-based chemically defined medium, the application of growing biocatalyst compared to resting whole-cells showed a higher volumetric product formation rate and biocatalyst activity (Process A). This may be attributed to increased cell metabolism in the growing system which favoured increased glucose uptake rate and biomass formation. However, the final product yield on catalyst was lower compared to the application of resting whole cells as biocatalyst. This may be due to glucose uptake for biomass formation which is usually at the expense of energy (ATP) and redox cofactors. Interestingly, the cells were able to increase and maintain P450 expression under biotransformation conditions until a drop in pH of the reaction medium below 6. This was apparently triggered by the formation of acetic acid (and possibly other acidic metabolites) which became inhibitory to the growing biocatalyst, preventing further formation of biomass. Beyond its effect on reducing medium pH to an unfavourable level, acetic acid also impacts synthesis of native and recombinant proteins (CYP153A6) which are essential in maintaining cell integrity for biotransformation. This unfavourable condition is possibly caused by overflow metabolism (of the carbon source) as oxygen depletion was discouraged by intermittent (6 to 8 hourly) opening of the vials to allow air inflow. In oxygenase catalysis, overflow metabolism constitutes energy drain thereby reducing NAD(P)H yield, causing metabolic shifts, uncoupling of proton gradient and ATP synthesis. Therefore, it is important to operate a system that discourages the formation of volatile fatty acids (e.g. acetate, formate and lactate), regardless of whether a buffered medium or pH controlled bioreactor is used. This can be achieved through a fed-batch operated process, which can also control biomass formation such that more cell energy and cofactor is available for biotransformation. Feeding strategies can be employed to control growth rate and attainable biomass concentration, specific productivity of recombinant proteins, and by product formation.

Table 7: Summary of results achieved under varied biotransformation conditions in processes A to E. This is limited to conditions under similar glucoamylase concentration (0.6 U L ^"1 ) and process temperature (20°C).

† - estimated, n.a. - not applicable, n.d. - not determined The novel EnBase ^® technology is an enzyme based system that releases glucose from starch polymer through glucoamylase activity (Panula-Perala et al., 2008). It was specifically designed for small scale agitated systems (such as shake flask and microtiter plate) but has been shown to be applicable in large scale bioreactors (Glazyrina et al., 2012, Siurkus et al., 2010). This system has been useful in improving biomass accumulation on small scale while also enhancing protein expression per unit biomass (Glazyrina et al., 2010, Hortsch and Weuster- Botz, 2011 , Siurkus and Neubauer, 201 1). However, it has not been tested under biotransformation conditions, particularly a two-phase biocatalytic reaction involving an organic phase. Glucoamylase concentration can be optimised to determine the preferred glucose release rate which in turn impacts the cell growth rate and biomass accumulation. The envisaged major advantage is the continuous and stoichiometric supply of glucose for energy and cofactor regeneration during biotransformation. It further ensures control of oxygen utilisation (a major reactant in oxygenase catalysis) to prevent anoxic conditions in the system which can also trigger acetic acid formation. In this study, the dependence of glucose release and cell specific growth rate on glucoamylase activity was demonstrated at different enzyme concentration. As previously reported (Panula-Perala et al, 2008), the glucose release (or flow) rate was a function of the corresponding glucoamylase concentration added to the reaction mixture. This allowed the enzyme concentration to be adjusted in the biotransformation studies to vary cell growth rate and facilitate process design. Surprisingly, the glucose release rate was not compromised in the presence of octane. This shows that the EnBase ^® system can function effectively in the two phase biocatalytic system involving octane as the substrate.

Medium composition has a substantial effect on the activities of whole cell biocatalyst in organic synthesis. To ensure comparable result with previously reported data using resting whole cells as biocatalyst in octane hydroxylation preliminary studies were carried out using resting whole cells cultured on EnBase ^® medium (Process B). In this study, optimum induction condition (22 h of cell cultivation, Table 1) was determined and used in biotransformation studies. It is necessary that optimum induction conditions are considered in the use of different growth medium as the latter impacts cell growth which, in turn, influences protein expression (Olaofe et al., 2010). Although there was a major increase in volumetric P450 concentration (1.67 μΜ), the specific enzyme content (CYP153A6) was similar to previously determined values with cells grown on LB and defined medium (0.15 - 0.20 μηιοΙρ45ο gDcw ^"1 ). However, the application of the resting whole cells showed higher activity (1.5 - 2.5 fold) compared to cells grown on LB and glucose-based defined medium. The biocatalyst efficiency of 1.93 g _oc tanoi gDcw ^"1 determined at 94 h, was higher than the efficiency achieved with cells grown on chemically defined medium in the same duration, 1.44 goctanoi gDcw ^" ■ This may be attributed to the consistent physiological status of the cells under defined conditions of energy provision.

The application of EnBase® grown cells resuspended in starch solution and used as resting biocatalyst under glucoamylase influence showed relatively lower activity and efficiency when compared to values achieved with glucose- supplemented biotransformation using cells grown on EnBase® (Table 7). This may indicate the preference of the cells for refined glucose relative to the one released from the starch polysaccharide.

Further biotransformation experiments were conducted with growing cells on EnBase® medium. The initial study was undertaken using protocol from BioSilta, a recommended glucoamylase concentration of 0.6 U L ^"1 and addition of booster additives for improved protein expression (Process C). While P450 concentration stayed the same at the start of biotransformation, 0.15 - 0.20 μηιοΙρ45ο gDcw ^"1 , a substantial increase in final product concentration (16 g _oc tanoi LBRM ^"1 , 2.2 g _oc tanoi gDcw ^{" 1} ) was observed. This can be ascribed to the addition of boosting solution containing amino acids, trace metals, cofactors and vitamins at the point of protein induction, shortly before the start of biotransformation. This is likely to have contributed to the significant increase in the final concentration achieved, which presently is significantly the highest concentration of 1-octanol reported in the literature. In subsequent experiments, biotransformations studies were carried out with EnBase® medium lacking complex nutrients. This was necessary to enable process control using glucose as the limiting substrate in the fed-batch operation.

In process D, the cells were induced at 14 h, incubated for another 2 - 3 h and used for biotransformation as the cells were at the exponential phase of growth. In the biotransformation studies lacking complex nutrients, the octanol accumulation over the initial phase of 9 h did not correspond to the initial glucose release rate at high amylase concentration. This indicates that some of the glucose were utilised for other cell functions including biomass formation and production of side products such as acetic acid. This became more obvious over longer duration of biotransformation (60 - 120 h), where the biocatalyst efficiency at lower amylase concentration (0.77 goctanoi gDcw ^" ) was almost 2 fold over the efficiency achieved at high glucoamylase concentration, 12 U L ¹ . However, this maximum value was lower than other values in process A to C. This may have been due to the induction of P450 expression at sub-optimal conditions (12 - 14 h), even though it was considered that exponentially growing cells may enhance octanol production.

Subsequently, the use of EnBase® Flo Zero was considered as it allows the cells to be resuspended in fresh medium lacking complex additives (process E). The cells were cultivated in EnBase® medium, induced at 22 h and incubated for another 3 h. The harvested biomass was resuspended in EnBase® Flo Zero to a concentration of 1 to 1.2 g L ¹ and supplemented with glucoamylase 0.6 to 12 U L ¹ to ensure the continuous supply of carbon for energy and cofactor regeneration. At low level of glucoamylase, octanol production was more efficient, reaching a maximum of 2.11 g _oc tanoi gDcw ^"1 in 120 h, a 3 fold increase over octanol achieved using 12 U L ¹ glucoamylase. As with studies in process D, reactions with higher concentration of glucoamylase (i.e. higher glucose release) produced more biomass and acetic acid (an indication of overflow metabolism). The latter resulted in acidification of the medium (i.e. lower pH) which became unfavourable to the system and inhibited the whole cell biocatalyst. These conditions serve as an energy (ATP) and cofactor (NAD(P)H) drain to the whole cell catalyst which would have otherwise been used to maintain cell integrity (i.e. membrane) and formation of the desired product in the biotransformation reaction. At lower level of glucoamylase (less than 0.6 U L ¹ ), glucose was released in limited fashion (Figure 9e) and appeared efficiently utilised under biotransformation conditions. The acetic acid excreted was considerably low (Figure 9c) while the pH of the system with 0.6 U L ^"1 glucoamylase was maintained between 6.4 and 7 over duration of 120 h (Figure 9d). It is also possible that increased cell growth rate under high glucose release contributed to acetic acid formation. While the growth rate in this study was less than 0.20 h ^"1 (Table 4), the coupled effect of bioconversion stress and relatively higher growth rate may have contributed to the substantial accumulation of acetic acid. The latter may have, in turn, influenced the overall carbon flux pattern and ATP metabolism. This may explain why the biocatalyst was less efficient at higher cell growth rate compared to the slow growing cells at reduced glucoamylase concentration.

In comparing results across the biotransformation conditions studied, processes A to E, the cells grown on EnBase® were more active and efficient in octane biotransformation, either as a resting or growing biocatalyst (Table 7). Although the growing cell was less efficient as biocatalyst in glucose-supplemented biotransformation under batch process, it fared better using EnBase® medium. The use of growing whole cells as biocatalyst in octane hydroxylation reaction showed 1.7 to 2 fold increase in biocatalyst efficiency and maximum activity compared to resting cells in the same medium. The biocatalyst efficiency and activity of the growing cells were enhanced under the influence of slowly released glucose in the range 0.04 to 0.08 g L ¹ hr ¹ , giving of 0.02 to 0.03 hr ¹ . The residual glucose concentration was maintained at ~ 0.5 g L ¹ through the bioconversion phase which discouraged overflow metabolism and relatively stabilised the reaction pH. However, this reduces biomass accumulation which is a requirement for increased volumetric concentration of octanol. This may be resolved in a controlled bioreactor environment where adequate nutrient and oxygen can be provided to improve biomass accumulation while maintaining a slow and linear growth rate under biotransformation conditions.

1.5. Conclusions

The application of growing whole cell as biocatalyst has not been considered in CYP153 based hydroxylation of alkane, possibly due to the enhanced cofactor pool under resting cell conditions. The application of resting cells as biocatalyst in the present study was shown to be over 50% more efficient in octanol formation compared with the use of growing cells in whole-cell based hydroxylation of octane, on a biomass basis.

As shown in this study, the use of growing cells as biocatalyst offers the potential of better process stability through maintaining the concentration of active P450. However, the biocatalyst efficiency is compromised due to formation of byproducts which serve as carbon sink and possibly as a result of limited cofactor availability. The use of EnBase® technology involving enzymatic glucose release to reduce carbon spilling and uncoupling of glucose utilisation for octanol production was applied. The slow and continuous supply of carbon for cell energy and cofactor regeneration during the growth phase under biotransformation conditions led to a 1.8 fold increase in biocatalyst activity compared with resting cells under identical reaction conditions. Surprisingly, the growing biocatalyst was more efficient by 40%. In the presence of boosting complex nutrients in the growing system, octanol concentration reached a maximum volumetric level twice that achieved using resting cells as biocatalyst.

The application of EnBase ^® (without complex additives) has shown the importance of slowly growing cells as a means of increasing the specific octanol yield and reducing carbon wasting in form of overflow metabolism and by-product formation.

EXAMPLE 2

2. Scale-up of biotransformation with whole-cell biocatalysts using a

bioreactor

2.1. Materials and Experimental Methods

2. 1. 1. Microorganism specifics

The recombinant Escherichia coli BL21 (DE3) strain (obtained from the Biocatalysis Laboratory, University of the Free State, South Africa). CYP153A6 was expressed by the insertion of a plasmid, pET28b(+), encoding the complete operon from Mycobacterium sp. HXN-1500 as well as the ferrodoxin reductase (FdR) and ferrodoxin (Fdx) redox partner proteins.

2. 1.2. Growth and biotransformation media

Pre-culture inoculum (50 mL in a 500 mL flask) was prepared by supplementing LB medium (10 g L ^"1 tryptone, 5 g L ^"1 , NaCI and 5 g L ^"1 yeast extract) with 30 ^yg mL ^"1 kanamycin and E. coli BL21 (DE3) pET28b-PFR1500 stock. The pre-culture was incubated at 30 °C and 160 rpm for at least 12 h before being used as an inoculum. Standard growth experiments were performed in EnBase® Flo (BioSilta) liquid growth medium (50 mL in a 500 mL baffled flask, 1.54 L in the stirred reactor (STR) or 0.924 L in the shaking reactor according to the recommended EnBase® protocol for growth and induction. The medium was inoculated with 2% (v/v) cells from the pre-culture inoculum (at an OD578 nm 5, 3.2 gDCW L 1) and the EnBase® components (0.6 μί ^"1 EnBase® enzyme (EnZ I'm), 0.1 g L ^"1 thiamine and 6.14 mM MgS04) as well as 30 μg mL ^"1 kanamycin were included at startup. Growth was carried out at 20 °C and 200 rpm for the vial apparatus, 860 rpm and 1 vvm for the STR or 175 rpm and 1.2 L total volume for the shaking reactor.

At induction, the growth medium was supplemented with 0.5 mM isopropyl β-D-l- thiogalactopyranoside (IPTG), 0.5 mM d-aminolevulinic acid (δ-ALA) and 50 μΜ FeC ^■ 6H2O as well as an additional 0.084 units EnZ I'm and 5 mL, 154 mL or 92.3 mL depending on the growth apparatus (10% of the working volume) EnBase® booster solution as recommended in the EnBase® Flo protocol. During cultivation samples were periodically taken for the determination of the growth rate and cell concentration.

2. 1.3. Analytical methods

2. 1.3. 1. Reactor sampling

10 mL reactor samples in 15 mL centrifuge tubes (Greiner) were centrifuged at 4000 rpm for 10 min in a centrifuge. The supernatant (including organic phase) was extracted and the remaining cell pellet washed by resuspension in 10 mL 0.9% (v/v) NaCL twice before biomass and CYP153A6 protein concentration anylses. The supernatant was subjected to solvent extraction before the aqueous and organic layers were separated.

2. 1.3.2. Biomass quantification

Biomass was quantified by optical density (OD) measurement at 578 nm. In addition, the dry cell mass was determined from a 2 ml sample of cell broth by drying the cell pellet after centrifugation at 80 °C for > 24 h, followed by desiccation in a desiccator to cool to room temperature before being weighed to four decimal places. 2. 1.3.3. CYP153A6 protein concentration analysis The amount of active CYP153A6 protein inside the cells was quantified using the CO-difference spectra method described by Guengerich et al. (2009). The volumetric concentration of P450, CP45O [nmolp45o mL ^-1 ] can be determined using an extinction coefficient of 91 mM ^"1 cm ^-1 as follows:

2. 1.3.4. 1-Octanol formation

In order to determine the concentration of 1 -octanol using gas chromatography (GC), the samples were subjected to solvent extraction. 5 mL ethyl acetate, containing 0.3% (v/v) 1-decanol as the internal standard was added to about 10 mL of collected supernatant including organic phase in a 50 mL centrifuge tube. The samples were mixed on an Intelli-Mixer at the highest setting for 10 min, followed by centrifugation at 4000 rpm for 3 min in a centrifuge. The organic layer was then carefully extracted (1 mL at a time) into a separate 15 mL centrifuge tube. Solvent extraction of the left-over supernatant was repeated with a further 5 mL ethyl acetate, followed by mixing, and centrifugation. The total organic extract was subjected to GC analysis.

2. 1.4. Oxygen transfer determination methods

2. 1.4. 1. Hydrogen peroxide method

Experiments were carried out at organic phase loadings of 0% and 23% (v/v) with remainder bulk solution being water. The temperature of the reactors was maintained at 20 ± 0.1 °C by ethylene glycol in the cooling jacket in all experiments and continuously checked with a thermometer. In the stirred reactor, volumetric aeration rates were varied from 0.5, 1 .0, 1 .5 and 2.0 vvm (corresponding to 1 .0, 2.0, 3.0 and 4.0 L min ^-1 respectively) at each agitation rate which varied from 360, 460, 580, 860 and 1360 rpm. In addition, the rate of passive diffusion of oxygen from the headspace was quantified at a zero aeration rate.

The orbital shaking reactor was characterized with and without the helical track modification at shaking frequencies of 150, 175 and 200 rpm and working volumes of 40%, 50% and 60% (corresponding to 0.8, 1.0 and 1.2 L respectively). In all instances when working with the shaking reactors, the aeration rate into the headspace was kept minimal at 0.5 L min ^-1 to mimic passive diffusion of air into the sealed vessel.

Dissolved oxygen tension (DOT) measurements were attained using an InPro R 6110 Series polarographic O2 sensor (Mettler Toledo) fitted with a PTFE coated membrane (T-96 type). The data was recorded using the Bioflo 1 10 Fermentor/Bioreactor control system (New Brunswick Scientific). The procedure as described by Vasconcelos et al., 1997 was adapted for use.

2. 4.2. (Dynamic) Gassing out method

The gassing out method for K/_a determination was also employed in both reactor setups in both the presence (23% v/v) and absence of an organic phase. The same operating conditions described in the hydrogen peroxide method (agitation and aeration) were used for the gassing out method.

The Oxygen Utilisation Rate (OUR) of the growing culture was established using the dynamic gassing out method. The protocol is the same as that of the gassing out method, but instead of sparging the medium with nitrogen, aeration was interrupted and the culture slowly used up the residual oxygen. The rate at which this oxygen was depleted is the OUR. The DOT was maintained above 30% air saturation to prevent the culture from becoming oxygen limited.

2.2. Results and Discussion

2.2. 1. Oxygen transfer characterization in a reactor

The hydrogen peroxide method for Kia determination was employed in the stirred reactor setup. For verification, the gassing out method was also employed and the results of the two methods were compared.

The hydrogen peroxide method underestimated the Kia when compared to those results generated from the gassing out method under the same conditions. The reactor oxygen transfer characterizations were therefore determined with the gassing out method. For the sake of consistency and accuracy, results for Kua determination using the gassing out method are reported including the probe response lag constant.

2.2. 1. 1. Oxygen transfer characterization (OTR) in the stirred reactor (STR)

The OTR characteristics of the STR are shown in Figure 1 1 as a function of agitation rate, aeration rate and organic loading. To maintain consistency and allow comparison between the studies performed by Olaofe (2013) in this same system, the organic loading was kept at 23% (v/v). The OTR trends in the STR at this organic loading are shown in Figure 11 b. Variance in the data was elevated at higher values of Kia as a result of the error of regression introduced when including the probe response time. The effect of increasing the aeration rate above 1 vvm had little effect on the oxygen transfer of the system. This was most likely a result of flooding of the impeller causing no further mass transfer advantages. Despite having exceptional Kia values at the highest agitation rate, it was hypothesized that running a growth system at 1360 rpm would be unfavourable because of the high fluid shear stress imposed by the impeller. For this reason, the most practical operating conditions of the STR were considered to be 860 rpm and 1 vmm.

2.2. 7.2. Oxygen transfer characterization (OTR) in the Orbital Shaking Reactor

The oxygen transfer characterization of the shaking reactor is presented in Figure 12. Previous oxygen transfer studies using shaking reactors by Zhang et al. (2008, 2009) were performed with open top vessels (passive diffusion of oxygen from air). In order to generate comparable results, the aeration rate in this study was maintained at a nominal 0.5 Lmin ^-1 (corresponding to 0.42-0.63 vvm at the various filling volumes) in order to mimic passive diffusion within the sealed reactor vessel.

The presence of a helical track modification on the oxygen transfer in a pure water environment is shown using a surface plot in Figure 12a. Up to 2-fold improvements in Kia can be observed when using a helical track. Interestingly, the highest Kia peaks are at different operating conditions depending on whether the helical track is utilized or not. The orbital shaking reactor without the helical track modification had a maximum Kia at the highest shaking frequency (200 rpm) and lowest filling volume (40%) whereas the reactor with a helical track favoured the largest filling volume (60%) at the highest shaking frequency. In the presence of organic (Figure 12b), the increase in Kia due to the helical track was not as prominent, with the majority of the data lying within the 95% confidence interval. This was most likely a result of the almost 2-fold Kia repression at an organic phase fraction of 23% (v/v) (Figure 12c) as well as the increased variance in the data observable in the presence of an organic volume.

Unlike the STR apparatus, the effectiveness of shaking reactors is more dependent on the physical reactor characteristics such as reactor geometry and filling volume. The OTR was much lower in the shaking reactor than in the STR, however still in a reasonable range for growth ( /_a > 5 h ^"1 ). It should be noted that the oxygen sensor placement in the shaking reactor (at the center of the reactor vessel) would pose the worst-case scenario for oxygen transfer as it was in the least homogenous region of the reactor, at the center of the vortex. For this reason, the effective Kia values of the shaking reactor may be underestimated when compared to the STR.

Shaking reactors are heavily influenced by scale. As is evident from Table 8, lower specific power inputs are required at larger filling volumes. The shaking reactor also had an overall lower specific power requirement than the STR.

Table 8: S ecific power inputs of the shaking reactor and STR.

2.2.2. Bioreactor operations

Optimal operating conditions for the stirred tank reactor (1 vvm, 860 rpm) and shaking reactor (175 rpm, 60% (v/v) filling volume with a helical track) were used. The EnBase® Flo medium within the reactors was inoculated at 2% (v/v) mid- exponential phase inoculum. Induction was performed at 24 hours post-inoculation and biotransformation was performed at 12 hours after induction. Each bioreactor experimental run was repeated such that the reproducibility of results could be established. The product formation results of each run are presented in Figure 13 with the overall results tabulated in Table 9. Error was quantified in the dry cell weight measurements by processing 3 aliquots of each sample.

The biotransformation trends of the bioreactors exhibited one major similarity: the orbital shaking reactor consistently had lower product formation rates and final product titres. This result was most likely a ramification of the lower overall volumetric oxygen mass transfer constants of the shaking reactor (with and without the helical track) when compared to the STR. Nevertheless, it was evident that the growth and biotransformation of the cultures in the shaking reactor were not oxygen limited. This may be concluded from the specific octanol formation trends in which biotransformations were sustained for the full 72 h (even at the highest cell densities) before slowly coming to an end. The shaking reactor had improved growth rates pre-biotransformation in all instances, despite having a lower OTR. Further, improved expression of P450 of cultures in the shaking reactor was observed in all instances (Table 9). It follows that, because the shaking reactor was not oxygen limited, growth in the STR was limited by factors other than oxygen transfer. It was postulated that shear imposed by the high agitation rate was the key criterion in explaining the reduced growth trends in the STR. Foaming in the STR posed an unexpected problem during operation such that the aeration rate had to be decreased to 0.5 vvm. Table 9: Bioreactor results from each repeat experimental

ax : ¾¾ ¾¾«¾t »y.w 6.?0

Mas Spesife Bloe^aiyst- Mm

iM tM: 0.97 m

¾te |¾^,: ·Γ ^: Γ ^!

5.00 5,76 , ^* )9

Activity

!/.<fiK> . _sr; ,. _r ¾ _X ;; _W - ^! - tR;i!-

Further, the pH control of the reactors was more refined in the second run. pH control via base addition was applied to the STR as only a decrease in pH was expected with the formation of any acetic acid. It was observed that after the addition of booster solution at the point of induction, the pH increased to values above 8 (in the first experimental run). The high nitrogen content of the booster solution may have caused ammonium-based product formation compensations from the cells, causing the pH of the medium to increase. The high pH was remedied by the manual addition of 30% H ₃ P0 ₄ (Panke et al., 2002). Manual pH control was performed in the shaking reactor at each sample time, using the STR control patterns as a guide. Any differences in results between first and second experimental runs may be attributed to this inadequate pH control. Specifically, the up to 3-fold improvements in protein expression exhibited in the second run was testament to the high dependence of CYP expression on pH. In the second experimental run, the refined pH control allowed biotransformation in the STR to proceed for longer, resulting in larger 1-octanol titres and maximum cell densities (Table 9).

It is worth noting that although the shaking reactor produced an over 2-fold higher enzyme concentration than the STR, product formation titres and rates were still lower. Thus, the system productivity does not solely depend on the amount of protein within the cell, but rather on other process condition variables. Interestingly, the pH of the medium did not significantly decrease which indicates minimal acetic acid production from a well-optimized fed-batch scheme made possible by the EnBase® medium.

Although larger final product titres, cell densities and protein concentrations were observed in the second experimental run, the maximum specific biocatalyst activities were higher in the first experimental run (and that of the shaking reactor was the highest measured activity at 20 °C for this system to date). As the cultures were not oxygen limited, there must have been another unmeasured factor at play.

Lastly, the results of this study are compared to that of previous studies in Table 10. Results from the STR and shaking reactor present the highest P450 enzyme concentrations (with that of the shaking reactor being over 2-fold higher than previously reported values) as well as the highest biocatalyst effiencies to date.

Other variables such as (specific) octanol concentration, volumetric product formation rate and biocatalyst activity were on par with previous studies (and more so for the STR than shaking reactor). The operation of the STR may be improved by addressing the foaming, shear forces and pH control. Although the shaking reactor did not perform as well as the STR, it can still be viewed as a successful alternative to conventional STRs when the above- mentioned issues are prevalent, offering a more convenient and simpler operation.

These results conclusively prove that the hydroxylation of n-octane using CYP153A6 expressed in E. coli can be effectively carried out in a reactor system. Table 10: Comparison with previous studies of 1-octanol production using CYP153A6 expressed in E. coli BL21(DE3) using a pET28b plasmid.

1.3 x HT* 10 ⁵ 1.3 i " ^i¾ i

2(5 m w m

11 5.4 3.7 7.8 MI . m ^■ ± &w la V i ⁽ Mmi $.a M S.9 4,5

Me Q idP^i km 41¾ 0.3ft mi m

Table lOcont.: Comparison with previous studies of 1-octanol production using CYP153A6 expressed in E. coli

BL21(DE3) using a pET28b plasmid.

0.V3 /A JS 15$

¾1> si Ornish rA i<m

M; A N/A a 12 0.28 ± 0.0 ί i: 0.01

Α·.:ξΐνί!.ν

Dwr^ii¾¾:M«¾ 3-24 &24 (MS 0-24 0?¾4 0-12 ⁽ 2

REFERENCES

Cornelissen, S., Julsing, M. K., Volmer, J., Riechert, O., Schmid, A. and Buhler, B. (2013). Whole cell based CYP153A6 catalyzed (S)-limonene hydroxylation efficiency depends on host background and profits from monoterpene uptake via AlkL. Biotechnology and Bioengineering, 1 10, 1282-1292.

Duetz, W. A., van Beilen, J. B. and Witholt, B. (2001). Using proteins in their natural environment: potential and limitations of microbial whole-cell hydroxylations in applied catalysis. Current Opinion in Biotechnology, 12, 419-425.

Glazyrina, J., Materna, E.-M., Dreher, T., Storm, D., Junne, S., Adams, T., Greller, G. and Neubauer, P. (2010). High cell density cultivation and recombinant protein production with Escherichia coli in a rocking-motion-type bioreactor. Microbial Cell Factories, 9, 42.

Glazyrina, J., Krause, M., Junne, S., Glauche, F., Strom, D. and Neubauer, P. (2012). Glucose-limited high cell density cultivations from small to pilot plant scale using an enzyme-controlled glucose delivery system. New Biotechnology, 29, 235- 242.

Hortsch, R. and Weuster-Botz, D. (201 1). Growth and recombinat protein expression with Escherichia coli in different batch cultivation media. Applied Microbiology and Biotechnology, 90, 69-76.

Krause, M., Ukkonen, K., Haataja, T., Ruottinen, M., Glumoff, T., Neubauer, A., Neubauer, P. and Vasala, A. (2010). A novel fed-batch based cultivation method provides high cell-density and improves yield of soluble recombinant proteins in shaken cultures. Microbial Cell Factories, 9, 1-11.

Olaofe, O. A., Burton, S. G., Cowan, D. and Harrison, S. T. L. (2010). Improving the production of a thermostable amidase through optimising IPTG induction in a highly dense culture of recombinant Escherichia coli. Biochemical Engineering Journal, 52, 19-24. Olaofe, O. A., Fenner, C. J., Gudiminchi, R., Smit, M. S. and Harrison, S. T. (2013). The influence of microbial physiology on biocatalyst activity and efficiency in the terminal hydroxylation of n-octane using Escherichia coli expressing the alkane hydroxylase, CYP153A6. Microbial Cell Factories, 12:8.

Panula-Perala, J., Siurkus, J., Vasala, A., Wilmanowski, R., Casteleijn, M. G. and Neubauer, P. (2008). Enzyme controlled glucose auto-delivery for high cell density cultivations in microplates and shake flasks. Microbial Cell Factories, 7:31.

Siurkus, J., Panula-Perala, J., Uwe, H., Rimseliene, R. and Neubauer, P. (2010). Novel approach of high cell density recombinant bioprocess development: Optimisation and scale-up from microtitre to pilot scales while maintaining the fed- batch cultivation mode of E. coli cultures. Microbial Cell Factories, 9, 35.

Siurkus, J. and Neubauer, P. (201 1). Heterologous production of active ribonuclease inhibitor in Escherichia coli by redox state control and chaperonin coexpression. Microbial Cell Factories, 10, 65.

Vallon, T., Glemser, M., Honda Malca, S., Scheps, D., Schmid, J., Siemann- Herzberg, M., Hauer, B. and Takors, R. (2013). Production of 1-Octanol from n- Octane by Pseudomonas putida KT2440. Chemie Ingenieur Technik, 85, 841-848

Yun, H. S., Hong, J. and Lim, H. C. (1996). Regulation of ribosome synthesis in Escherichia coli: Effects of temperature and dilution rate changes. Biotechnology and Bioengineering, 52, 615-624.

Zhang, X., Burki, C.-A., Stettler, M., De Sanctis, D., Perrone, M., Discacciati, M., Parolini, N., DeJesus, M., Hacker, D. L, Quarteroni, A., and Wurm, F. M. 2009. Efficient oxygen transfer by surface aeration in shaken cylindrical containers for mammalian cell cultivation at volumetric scales up to 1000L. Biochemical Engineering Journal, 45:41-47.

Zhang, X., Stettler, M., Reif, O., Kocourek, A., Dejesus, M., Hacker, D. L, and Wurm, F. M. 2008. Shaken helical track bioreactors: Providing oxygen to high density cultures of mammalian cells at volumes up to 1000 L by surface aeration with air. New Biotechnology, 25(1):68-75.